Looking for advice on handling 503 Service Unavailable errors when simulating high-concurrency webchat sessions.
Environment: Genesys Cloud US1
Tool: JMeter 5.6
Endpoint: /api/v2/conversations/messaging
We are running a load test to validate the messaging capacity for a new digital channel deployment. The goal is to simulate 200 concurrent users initiating webchat sessions via the Architect flow. The test setup involves sending POST requests to create messaging conversations with a pre-configured participant.
The test fails after approximately 45 seconds. The JMeter results show a spike in 503 errors when the concurrent thread count exceeds 100. Before this, the requests return 200 OK. We have checked the API rate limits documentation, but we are not hitting the 429 Too Many Requests limit. The error seems to be related to backend service availability rather than rate limiting.
We are using the following payload for the conversation creation:
{
"type": "webchat",
"to": {
"id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
"type": "user"
},
"from": {
"id": "user-session-12345",
"type": "user"
},
"labels": ["load-test"]
}
The Architect flow is configured to route these messages to a specific queue. The queue has 50 agents available. We are not seeing any errors in the agent scripts or routing logic. The issue appears isolated to the initial conversation creation API call.
Is there a specific limit on the number of concurrent messaging conversations that can be created per second? Or is this a known issue with the webchat channel during peak load? We need to determine if this is a configuration error in our test or a platform limitation. Any guidance on how to adjust the test parameters or diagnose the root cause would be appreciated. We are currently blocked on validating the digital channel capacity.